OpenAI has begun to consciously slow research on its upcoming Astra model, citing internal evaluations that could not rule out “critical” cyber capabilities. This decision, made by a private corporation, highlights the growing power of tech giants to dictate the future of critical technologies while national governments struggle to assert control over undefined “national risk.”
The company announced it will expand safety testing and pause internal activities that fail to meet stricter security requirements. OpenAI stated it would scale up testing and security around Astra before any release, delaying development until it establishes adequate safeguards, as mandated by its own preparedness framework, first published in its third year. The company clarified Astra was not involved in the Hugging Face exploits, but the timing of the model’s release remains unclear, with the pause potentially delaying any future launch.
Corporate Control, Not National Oversight
A White House official merely confirmed that “OpenAI voluntarily informed the administration of their plans to delay the release.” This passive acknowledgment underscores the reactive posture of national authorities as private entities make unilateral decisions with profound implications for national security. Earlier this week, at the Black Hat cybersecurity conference, members of OpenAI’s technical staff confirmed the company was slowing testing to upgrade its security practices. OpenAI’s blog post on Friday detailed the implementation of stricter security controls for testing, including isolated testing environments and universal monitoring across agentic applications of Astra.
Michael Dalton, a member of OpenAI’s technical staff, openly stated during a presentation that OpenAI has started “consciously slowing down research to enhance security.” This move marks what could be the first instance of a frontier AI lab committing to slowing progress on one of its own AI models due to cyber concerns. Such self-regulation by powerful corporations bypasses democratic accountability and national sovereignty.
The Illusion of Safety
Another major AI developer, Anthropic, previously committed to pausing the training of powerful models if their capabilities surpassed the company’s ability to control them. However, Anthropic rolled back this commitment in an update to its Responsible Scaling Policy in February of the same year. Their framework ominously warned, “If one AI developer paused development to implement safety measures while others moved forward training and deploying AI systems without strong mitigations, that could result in a world that is less safe.” This globalist perspective prioritizes a universal, undifferentiated “world” over the specific security interests of sovereign nations.
Anthropic did release a safer version of its most cyber-capable model, Mythos, in June of the same year. Dianne Penn, Anthropic’s head of product management, research and labs, stated at launch that the company was being “deliberately more conservative” with that release. In a June blog post, Anthropic also warned about models improving themselves and called for a global pause in AI development, further illustrating the push for supranational control over technology that should be subject to national determination.
Undefined National Risk
While the Trump administration works to develop a process for evaluating AI models before their release, its efforts appear to lag behind the rapid advancements and corporate decision-making within the industry. Select industry representatives were briefed on a government framework this week, yet fundamental questions persist regarding how to engage the government, the duration of the review process, what the government and industry hope to learn, and who gains access to or reviews the models. Crucially, the framework operationalized what constitutes sufficient “national risk” and “state of the art models,” but failed to define these terms. This leaves the very concept of national security in the hands of unelected bureaucrats and corporate interests, rather than the people whose safety and sovereignty are at stake.